Related Experiment Video
Updated: Aug 19, 2026
Establishing a Competing Risk Regression Nomogram Model for Survival Data
Published on: October 23, 2020
Development and Validation of a Nomogram for Predicting Mortality in Patients with Multidrug-Resistant Bacteria
Yahan Li1, Xiaorong Lai2, Yaping Yao1
1Department of Disease Prevention and Control, Air Force Hospital of Eastern Theater, Nanjing, 210002, People's Republic of China.
Background:
Multidrug-resistant bacterial (MDRB) infections are a serious public health threat in the era of widespread antibiotic misuse.
Objective:
To construct and validate a mortality risk prediction model for MDRB infection.
Materials And Methods:
This retrospective cohort study was conducted at the Air Force Hospital of the Eastern Theater Command between January 2019 and December 2025. A total of 1, 071 patients infected with MDRB were divided into two groups according to treatment outcomes: successful treatment and normal discharge (NG, n = 991) and treatment failure leading to death (DG, n = 80). Cox proportional hazards regression analysis was performed to determine the independent risk factors associated with MDRB infection. In addition, environmental samples were collected from hospital beds in three departments, and bacterial colonies were counted and species identified. A nomogram was developed based on these risk factors. Model performance was evaluated using a calibration plot, receiver operating characteristic (ROC) curve, and decision curve analysis (DCA).
Results:
Age at hospital admission, number of hospital discharge diagnoses, current bed location, duration of antimicrobial use, number of ventilator, number of central venous cannulations, number of catheter insertions, and whether the first surgery included postoperative prophylactic antimicrobial therapy were independent risk factors for mortality in the DG (all p < 0.05). The calibration plot and DCA demonstrated that the model exhibited good calibration and clinical utility. Current bed location was identified as an independent predictor of mortality in MDRB infection based on Cox analysis.
Conclusion:
The risk nomogram model is effective in predicting mortality risk in patients with MDRB infection and can be used to assess individual risk and guide preventive treatments and nursing interventions. The impact of environmental microbiology on patient bed locations cannot be ignored, and the prognostic significance of this factor has been observed in hospitalized patients with MDRB infection.